bme680 vs. ccs811

Good question. I havent any experience with either. I will need to get something of this sort to monitor gas in my RV... The project has not progressed that far yet..

What is the application you are looking at ?

@mark-bolton

Hi, Mark

Thank you for your response. I am looking at these two sensors on the Utmel website.  But I am looking forward to hearing someone's advice to decide which sensor is better to choose.

I also look the reference datasheet and parameters on the website.

You may see this.

Yours sincerely, 

Tia

Both of the Spec Sheets on your website are extensive and have very satisfying levels of detail.  The expression in English is very good too.  From the summary both seem to do what I would want and I will definitely consider them when it comes time to build the sensor package for my RV. 

The Gas monitoring facility is very important since RVs almost universal y use LPG and I have seen one explode from a fault.

The other parameters are interesting but not critical.

I haven't yet had time to decide what the differences between the two units is.

Since you are the Editor for the site and I assume English is a second language for you. If you would like someone to proof read you Product Descriptions for you, let me know.

Hi @tia,

  I haven't worked with any gas sensors, so all I can do is make some comments on the specification sheet data.

The common point of the two modules appears to be a volatile organic compound (VOC)  sensor that will detect molecules likely to be associated with everyday human activity, with examples mentioned including ethanol, acetone and carbon monoxide. Neither will detect carbon dioxide which is the most obvious contaminant in a poorly ventilated space inhabited by humans. It is presumed they will also detect many common contaminants from paints, etc., although the CCS811 datasheet is less forthcoming than the Bosch datasheet. It is generally unclear how the sensitivity varies between different contaminates for either sensor.

They also both incorporate a microprocessor to translate the raw resistance measurements. The quality, implicit assumptions and capability of this processing is unclear in both cases.

Hence, it is difficult to see how either sensor can be used, beyond a simple good/not good warning system based on each measurement. In a situation where one or a small number of specific VOCs are likely to occur, such as from a gas leak, then it would be necessary to do some calibration experiments involving those materials, but this process could result in a useful device for those specific applications.

The data explicitly provided does not enable a direct comparison of sensitivity of the two modules.

Both VOC sensors are affected by humidity and temperature. They both provide a means of adjusting the VOC reading to compensate, but only the Bosch module has temperature and humidity sensors built in. The CCS811 relies an on external data stream of temperature and humidity measurements, which is obviously much less convenient. It is conceivable that in some situations in which the range of temperature and humidity is quite small, that this compensation is not required, but this would need to be proven.

The CCS811 claims to produce an estimate of the carbon dioxide level, assuming some sort of 'normal human conditions'. The data sheet does not provide any evidence for this correlation. I would require substantial independent evidence to be convinced that this correlation is valid - I suspect the correlation is tenuous.

It is interesting to note that most of the uses suggested for the Bosch module, such as weather forecast or navigation enhancement, are unlikely to make use of the VOC sensor data.

---------

In summary, it is possible to envisage either device being used to monitor air for a gas leak or similar VOC 'leakage' situation, in which prior calibration experiments have been used to determine threshold limits.

Such an approach could be extended to the air in a room inhabited by humans, again following some calibration experiments, to warn of 'poorly ventilated' space or activate an air conditioning system. I presume it would be possible to calibrate either sensor to give a warning if the VOC sensor gives a measurement above a certain level, but I have my doubts as to how useful that would be as general measurement device.

Neither sensor can measure the level of a specific contaminant, which limits the extent to which they can accurately determine an air quality.

@mark-bolton

Hi, Mark.

I am very happy to see your reply. Thank you so much.

Be grateful to receive your advice. I will do that so.

@davee

Hello. DaveE.

Thank you for your reply. Your suggestion is very clear and makes sense. I also got some ideas to decide which to choose. Really appreciate it!

The application for both of these devices is more likely to be monitoring an environment for the risk of fire / explosion due to a buildup of flammable hydrocarbons.  That is certainly what I will be looking at when it comes time to build that module in my Smart Homes system. 

In this instance what one would do is set the device up so as it wont trigger false positives in the absence of contaminants but will trigger as early as possible / lowest threshold in the event of a gas leak or flammable liquid spill.  In such an application linearity or calibration are not that important.

Robust reliability, resistance to sensor "poisoning"  and stability are important. 

I haven't checked on Cost but suggest that that wouldn't be a major factor in choosing between either device  since the cost in circuit design time and effort and the "mission critical" nature of the system out weigh dollar cost considerations.

M

@mark-bolton

Hi Mark,

   I guess you are looking for butane or propane. As you say, these can indeed cause major fires and explosions, particularly if they can leak into 'dead spaces' which do not vent downwards, as unlike lighter gases (e.g. hydrogen), they are heavier than air and can collect towards the bottom of a vehicle.

Whilst I would not automatically rule out either of the devices discussed, they appear to be designed for other purposes. I would imagine there are other devices designed for this particular type of application that might be more appropriate, as it is a common problem.

In addition, I am sure you already have in mind, but just in case, you should design should limit the power going to the sensors, etc., which are in the area that the gas is liable to collect, to a very low power level. The concern being, if a connection becomes faulty, etc, the power levels should intrinsically be less than those needed for ignition by a spark or hot point (inside and outside of the sensor). There may be regulations,  requirements or recommendations specific to RVs. If not, maybe check in other similar situations, such as boats and aircraft.

Good luck and take care, especially when you come to test your design! Best wishes, Dave

I am preparing to build this module now. Am I right in thinking that only the CCS811 has Carbon Monoxide sensing? Though it is technically frowned upon (possible even illegal) I have my instantaneous gas hot hater system indoors. If one were to close the windows and run it for some time I believe it would be possible to asphyxiate.   The BME680 looks a little less complex to implement. I am leaning towards the former.

Hi @mark-bolton,

The BME 680 data sheet includes:

Although, I am not clear how that sensitivity compares to a carbon monoxide detector provided for safety reasons.

Either way, I haven't seen a gas water heater that isn't of the balanced flue variety in the last 50 years...and that includes caravans (and I presume RVs) .... In the 1950s, there used to be a lot of them in kitchens and bathrooms, but they killed too many people!

That is, both the incoming air and outgoing exhaust should directly vent to the outside of a building, with the gas flames sealed away from the air internal to the building.

Personally, as well as exchanging any 'dodgy' water heaters as a first priority, if carbon monoxide build up is feasible, I would back any 'home made' detector up with at least an 'off the shelf' battery powered detector, which are available quite cheaply.

Keep safe, my friend!!

It is likely that the configuration of a LPG gas burning water heater right next to a shower recess in an RV would horrify some.  The weather is warm enough that we open the doors when we shower so CO poisoning is something we avoid habitually, rather than through any quantitative Science.  Sure I am breathing lots of CO ...

If it were really cold and I were really muddy I could see me just shutting out all ventilation and luxuriating under the hot water until forever...  ... but who knows if the CO would put me down for a Darwin Award ?

It would be interesting to quantify it.

Kindest

M

Hi @mark-bolton,

   I have the impression that Darwin's theorem is usually summarised as "survival of the fittest" rather than "elimination of the foolhardy", so presumably only the survivors will qualify for a Darwin award? 🤔 🤔 

I don't personally know the specific carbon monoxide levels that are dangerous, but I note:

• https://www.cpsc.gov/Safety-Education/Safety-Education-Centers/Carbon-Monoxide-Information-Center/Carbon-Monoxide-Questions-and-Answers .. which includes
  • "CO alarms are available for boats and recreational vehicles and should be used. The Recreation Vehicle Industry Association requires CO alarms in motor homes and in trailers."
• Wikipedia : https://en.wikipedia.org/wiki/Carbon_monoxide_poisoning    .. which includes
  • "Carbon monoxide poisoning is relatively common, resulting in more than 20,000 emergency room visits a year in the United States. It is the most common type of fatal poisoning in many countries. In the United States, non-fire related cases result in more than 400 deaths a year.

  • World Health Organization recommendations

    The following guideline values (ppm values rounded) and periods of time-weighted average exposures have been determined in such a way that the carboxyhaemoglobin (COHb) level of 2.5% is not exceeded, even when a normal subject engages in light or moderate exercise:

    • 100 mg/m3 (87 ppm) for 15 min
    • 60 mg/m3 (52 ppm) for 30 min
    • 30 mg/m3 (26 ppm) for 1 h
    • 10 mg/m3 (9 ppm) for 8 h
    • 7 mg/m3 (6 ppm) for 24 h (for indoor air quality, so as not to exceed 2% COHb for chronic exposure)

I note the lower levels (6-9 ppm) are lower than that mentioned in the Bosch sensor data sheet of 15 ppm, though I don't understand understand the Bosch data sufficiently to comment on the implications.

.........

  Take care my friend ... please don't take unneccesary risks... Wikipedia article points to carbon monoxide can causing long term illness as well as fatality.

@mark-bolton
How about a solar hot water system?

@davee
Ultimately evolution is about reproductive success and the ability to generate variations for natural selection to act upon.

Hi @robotbuilder,

  "Ultimately evolution is about reproductive success and the ability to generate variations for natural selection to act upon."

 I am not a biologist and I think this may heading far from the point of comparing sensors, but to my simplistic understanding, although reproductive success and resultant variations are obviously essential ingredients to Darwin's theory, surely the point is that some of the resulting variations only become more dominant (successful) when those respective variations bestow a survival advantage in the environment.

I doubt if any "advantages" @mark-bolton might acquire from risking carbon monoxide inhalation will offset the only too well-known disadvantages.

Solar heating is indeed worth considering ... when the weather permits.

Best wishes and take care to you both.